Notice: Undefined index: PATH_INFO in /home/u842710795/domains/sciencworldpublishing.org/public_html/abstract.php on line 3
Comparison of Count Models and associated risk factors of Neonatal Mortality in Ethiopia 1

Clinical Biostatistics and Biometrics

Comparison of Count Models and associated risk factors of Neonatal Mortality in Ethiopia

Million Wesenu Demissie

 

Department of Statistics, College of Computing and Informatics, Haramaya University, Dire Dawa, Ethiopia

 

*Corresponding author: Million Wesenu Demissie, Department of Statistics, College of Computing and Informatics, Haramaya University, Dire Dawa, Ethiopia, E-mail: millionwesu@gmail.com

 

Citation: Million Wesenu Demissie, (2020) Comparison  of Count Models and associated risk factors of Neonatal Mortality in Ethiopia. Clin Biostat Biometr. 1(1);1-9

 

Copyright: Â© 2020, Million WD, This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.

 

ABSTRACT

 

The Duration from 0- 28 day of life is a crucial time for survival of any child. Reducing neonatal mortality rate is a serious problem in many low and middle-income countries including Ethiopia. The main objective of this investigation was to find out the associated risk factors for number of neonatal mortality per women and select models that best fit of the neonate dataset. Recorded survey data of Ethiopian Demographic health survey in 2016 were obtained for a retrospective of 9958 women aged 15-49 years and considered in this investigation. Count models of  Standard Poisson, zero-inflated Poisson , Negative binomial Poisson and zero-inflated negative binomial regression were used to distinguish the potential risk factors and model comparison have been done using Bayesian Information Criteria (BIC) and Akaike Information Criteria (AIC). Based on this, zero-inflated negative binomial regression model had minimum value of AIC and BIC when compared with other models and best fit for the neonatal mortality dataset. From  results of zero-inflated negative binomial regression model parental Residence, Geographical-Region, education level of mother, Religion, mother Age at first time of birth, Source of drinking  water, Antenatal care visits, Husband/partner’s education level, postnatal care visit and Toilet facility were found to be statistically significant factors of neonatal mortality per mother.  It is therefore recommended that women should be attending antenatal care and postnatal care visit during pregnancy and after birth, respectively in addition to improve level of education for the better survival neonates in the first 28th day of life.

 

KEYWORDS: Count model, Over-dispersion, Neonatal, Akaike information criteria

Indexing Partners

image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing
image-missing